Antiviral Research 47 2000 1 – 17
Review
Resistance of influenza viruses to neuraminidase inhibitors — a review
Jennifer L. McKimm-Breschkin
Biomolecular Research Institute,
343
Royal Parade, Park6ille
3052
, Australia Received 21 March 2000; accepted 9 May 2000
Keywords
:
Influenza virus; Neuraminidase inhibitors; Resistance to neuraminidase inhibitors www.elsevier.comlocateantiviral
1. Introduction
Influenza virus is a negative stranded RNA virus. It contains two surface glycoproteins,
hemagglutinin HA and neuraminidase NA. HA exists as a trimer and is responsible for
binding to the terminal sialic acid bound to recep- tors on the surface of the target cell, leading to
attachment and subsequent penetration by the virus into the cell. Influenza virus isolates from
different animals appear to have a preference for specific receptor linkages. Equine and avian iso-
lates bind preferentially to the a2,3 galactose structure, while human isolates bind preferentially
to the a2,6 galactose structure Leigh et al., 1995. A single amino acid mutation is sufficient to
change receptor specificity Rogers and Paulson, 1983; Nobusawa and Nakajima, 1988; Martin et
al., 1998. Sequence analysis and alignment has identified key residues conserved across all HA
subtypes, which are involved in receptor binding Nobusawa et al., 1991. The sialic acid binding
site forms a groove across the top of the HA surrounded by antibody binding sites Weis et al.,
1988. Residues 134 – 138 form the right side of the ligand binding site, and residues 224 – 228
form the left side. Other conserved residues ap- pear to play a role in orienting several of the
surface atoms for binding to the sialic acid, these include Tyr 98, Trp 153, His 183, Glu 190, Leu
194 and Tyr at 195 Weis et al., 1988; Nobusawa et al., 1991.
After replication of the virus, progeny virions bud from the cell surface. NA is thought to be
responsible for cleavage of terminal sialic acid moieties from receptors, to facilitate elution of
progeny virions from the infected cell. Since they are also glycosylated, newly synthesized HA and
NA on virions may also contain sialic acid residues on their oligosaccharide chains. Removal
of these terminal sugars is therefore also necessary to prevent self-aggregation, due to the HA of one
virion binding to the sialic acids on an adjacent
Fax: + 61-3-96627101. E-mail address
:
jennybbioresi.com.au J.L. McKimm- Breschkin.
0166-354200 - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 3 5 4 2 0 0 0 0 1 0 3 - 0
virion. NA may also play a role in enabling the virus to penetrate the mucin layer in the respira-
tory tract. Structural analysis of NA of influenza virus
Varghese et al., 1983 revealed it was a tetramer, and that there were several invariant residues in
both influenza A and B which formed part of the active site. There are nine conserved residues
which contact the sialic acid, and there are a further ten which provide substructure or a scaf-
fold, on which amino acids contacting the bound sugar are supported Burmeister et al., 1992; Var-
ghese et al., 1992. The conserved residues include three arginines at positions 118, 292 and 371,
which interact with the carboxylate of the sugar. Other important residues include the hydroxyl
groups of the glycerol side chain which are hydro- gen bonded to Glu 276 and the 4-hydroxyl group
on the sugar is directed towards Glu 119 Vargh- ese et al., 1992. Based on the structure of NA in
complex with sialic acid Varghese et al., 1992 a program of inhibitor design was begun, based on
the earlier work of Meindl and Tuppy who first synthesized the unsaturated sialic acid analog,
Neu5Ac2en
2-deoxy-2,3-didehydro-
D
-N-acetyl- neuraminic acid or DANA Meindl and Tuppy,
1969. DANA is a weak inhibitor of NA enzyme activity and demonstrates weak antiviral activity
in vitro, but not in vivo when administered in- traperitoneally Palese and Schulman, 1977. Two
new inhibitors were designed, substituted at the 4 position, 4-amino-Neu5Ac2en, and 4-guanidino-
Neu5Ac2en, zanamivir, von Itzstein et al., 1993. Zanamivir has a 100-fold higher affinity than the
4-amino-Neu5Ac2en, which in turn is around 100- fold more effective than DANA. The 4-amino
group is predicted to form a strong salt bridge with the acid group of Glu 119 and the 4-
guanidino group is predicted to interact not only with Glu 119, but also with Glu 227 von Itzstein
et al., 1993 accounting for the further increase in its affinity. Zanamivir is not just an effective
inhibitor of the viral enzyme, but is a potent inhibitor of virus replication in both cell culture
and animals von Itzstein et al., 1993; Woods et al., 1993. The inhibitor prevents release of
progeny virions from the infected cells, and in vivo is only effective when applied at the site of
infection, either by inhalation or intranasal instil- lation von Itzstein et al., 1993; Hayden et al.,
1996, 1997 since orally administered drug is rapidly cleared by the kidneys Woods et al.,
1993. Zanamivir has now been approved for general use in the USA, Europe, Australasia and
Japan.
Based on the efficacy of zanamivir, other neu- raminidase inhibitors are also being developed.
GS4071, oseltamivir carboxylate, the active form of the ethyl ester prodrug GS4104, oseltamivir
phosphate, is a potent carbocylic inhibitor with a cyclohexene scaffolding Li et al., 1998. In addi-
tion, oseltamivir contains a bulky lipophilic side chain, a pentyl ether, at the 6-position, instead of
the glycerol group in zanamivir, and is orally active when administered as the ethyl ester pro-
drug, oseltamivir phosphate, which undergoes ac- tivation by hepatic esterases. It has been approved
for marketing in the USA and Sweden. Johnson and Johnson are developing an orally active cy-
clopentane derivative from Biocryst Pharmaceuti- cals Bantia et al., 1999a,b which has both the
guanidinium group as well as a bulky hydropho- bic side chain in a position sterically correspond-
ing to the glycerol side chain in sialic acid.
However, as with any new anti-infective it is important to establish whether resistance to these
new inhibitors arises readily. There have now been several publications in this field, predomi-
nantly on the characterization of laboratory gen- erated mutants, unexpectedly the mechanisms of
resistance are complex. As the drugs are now available for clinical use the aim of this paper is to
review the current available data on resistance, to provide some insight into the challenge faced in
trying to evaluate clinical isolates for possible resistance.
2. Generation of mutants
